1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
|
<!-- header fragment for html documentation -->
<!--$Id: compiling.html,v 1.37 2013/11/08 23:09:53 mkkuhner Exp $ -->
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2 Final//EN">
<HTML>
<HEAD>
<META NAME="description" CONTENT="Estimation of population parameters using genetic data usi
ng a maximum likelihood approach with Metropolis-Hastings Monte Carlo Markov chain importanc
e sampling">
<META NAME="keywords" CONTENT="MCMC, Markov chain, Monte Carlo, Metropolis-Hastings, populat
ion, parameters, migration rate, population size, recombination rate, maximum likelihood">
<TITLE>LAMARC Documentation: Compiling LAMARC</title>
</HEAD>
<BODY BGCOLOR="#FFFFFF">
<!-- coalescent, coalescence, Markov chain Monte Carlo simulation, migration rate, effective
population size, recombination rate, maximum likelihood -->
<P>(<A HREF="mapping.html">Previous</A> | <A
HREF="index.html">Contents</A> | <A HREF="converter.html">Next</A>)</P>
<H2>Compiling LAMARC</H2>
<P> We provide executables of LAMARC for many systems, but if you cannot use
these, or wish to compile the program yourself, we provide source code as well.
This article discusses our experience compiling LAMARC. We welcome
your comments on the use of other compilers and computing environments.</P>
<P> LAMARC is written in C++. We tried to avoid advanced or new features of
the language, but very old compilers will probably not work. We
recommend the use of GNU g++ on any machine which supports it; this is
the compiler we used to develop LAMARC, and it works well on most systems.</P>
<h3>Compilation and Testing of Released Executables</h3>
<P> Here is a list of machine/compiler combinations we have tried, with
comments on our success or lack of it.</P>
<TABLE BORDER="3" CELLSPACING="1" CELLPADDING="1">
<TR>
<TD width=13%><b>OS Family</b></TD>
<TD ><b>Compiler Used</b></TD>
<TD width=13%><b>Compiled On</b></TD>
<TD width=13%><b>Tested On</b></TD>
<TD width=39%><b>Notes</b></TD>
</TR>
<TR>
<TD rowspan=4>Linux</TD>
<TD>GNU g++ 4.5.1</TD>
<!-- lamarc bleeding edge -->
<TD colspan=2>red hat<br>2.6.18-128.1.1.el5<br>x86_64</TD>
<TD rowspan=4>
<!-- add info on cluster and special systems -->
Our code should compile on any Linux- or Unix-based system.
If you are using a compiler other than g++, the GNU C++
compiler, you may need to <a href="#advanced">give our configure script
and makefiles some extra help</a>
<br>
<br>
Previous releases have successfully compiled on 32 bit red hat
machines, but we did not have one at our disposal for this
release.
</TD>
</TR>
<TR><TD>GNU g++ 4.3.3<br><em>(released 64-bit executables)</em></TD>
<!-- lamarc modules -->
<TD colspan=2>red hat<br>2.6.18-128.1.1.el5<br>x86_64</TD>
</TR>
<TR><TD>GNU g++ 4.1.2</TD>
<!-- lamarc old -->
<TD colspan=2>red hat<br>2.6.18-128.1.1.el5<br>x86_64</TD>
</TR>
<TR><TD>GNU g++ 4.3.2</TD>
<!-- kingman -->
<TD colspan=2>debian<br>2.6.18-4-k7<br>32 bit i686</TD></TR>
</TR>
<TR>
<TD rowspan=2>Windows</TD>
<TD>MinGW cross compile<br><br>
GNU g++ 3.4.2<br>
MinGW runtime 3.7 <br>
w32 API 3.2 <br>
<em>(released 32-bit lam_conv executable)</em>
</TD>
<TD>red hat<br>2.6.18-128.1.1.el5<br>x86_64</TD>
<!-- check the windows version -->
<TD rowspan=2>64-bit<br>Windows 7 Professional</TD>
<TD rowspan=2>
We create our Windows executables by cross compiling from
Linux and/or Windows using a Mingw-w64 toolchain
of g++.
</TR>
<TR><TD>MinGW64 cross compile<br><br>
GNU g++ 4.5.1<br>
i686 Cygwin 1.7.7<br>
<em>(released 64-bit lamarc executable)</em>
</TD>
<TD>64-bit<br>Windows 7 Professional</TD>
</TR>
<TR>
<TD rowspan=4>MacOS X</TD>
<TD rowspan=2>GNU g++ 4.0.1<br></TD>
<TD rowspan=2>32 bit intel-mac<br>
OS X 10.4.11<br>
Darwin Kernel 8.11.1</TD>
<TD>32 bit intel-mac<br>
OS X 10.4.11<br>
Darwin Kernel 8.11.1</TD>
<TD rowspan=4>
As of Lamarc version 2.0, we have stopped supporting the Mac OS 9 platform.
<br>
<br>
We no longer distribute 32-bit executables for the Mac
but are still able to compile them. If you have a 32-bit
machine, and cannot compile your
own executables, please contact us at
<A HREF="mailto:lamarc@u.washington.edu">lamarc@u.washington.edu</A>
and we'll see if we can get you an executable that works for you.
<br>
<br>
The distributed wxWidgets code will not compile with g++ 4.0.0.
<br>
<br>
<tt>lam_conv</tt>, the lamarc file converter does not compile on
Mac OS X 10.6 in its current form. However, <tt>lamarc</tt> will.
We recommend you use a copy of <tt>lam_conv</tt> provided in the
.dmg file and compile <tt>lamarc</tt> to run on your
newer Mac. The <tt>lamarc</tt> in the .dmg should also run on
your machine, but a natively compiled version may be faster.
</TD>
</TR>
<TR><TD>32 bit powerpc<br>
OS X 10.4.11<br>
Darwin Kernel 8.11.0</TD></TR>
<TR><TD rowspan=2>GNU g++ 4.0.1<br><em>(released 64-bit executables)</em></TD>
<!-- elizabeth's mac -->
<TD rowspan=2>64 bit intel-mac<br>
OS X 10.5.8<br>
Kernel 9.8.0</TD>
<TD>64 bit intel-mac<br>
OS X 10.5.8<br>
Kernel 9.8.0</TD></TR>
<TR><TD>64 bit intel-mac<br>
OS X 10.6.2<br>
Kernel 10.2.0</TD></TR>
</TABLE>
<h3><a name="basic">Basic Instructions for Compiling and Installing LAMARC on *-nix Systems</a></h3>
<p>
Here are basic instructions for compiling on Linux/Unix/Mac OSX
systems.
</p>
<p>
Begin by un-taring the distribution and creating a sub-directory
to compile in. This will keep your object files
from cluttering up a listing of the top level of the distribution.
<pre>
tar xfvz lamarc-2.1.9-src.tar.gz
cd lamarc-2.1.9
mkdir release
cd release
</pre>
</p>
<p>
The configure script queries your system and produces a
Makefile tailored to your computing environment. Try invoking it
like this:
<pre>
../configure
</pre>
or, if that doesn't work, like this:
<pre>
sh ../configure
</pre>
</p>
<p>
The configure script will probably complain that you don't have
<a href="http://www.wxwidgets.org">wxWidgets</a> installed.
(wxWidgets is a free, open source toolkit for GUI applications.
Through it, we provide native look-and-feel Linux, Mac OS X, and
Windows GUI interfaces to our file converter using a single code
base.)
If you already have wxWidgets, you can invoke <tt>configure</tt> like this:
<pre>
sh ../configure --with-wx-config=/path/to/wx-config
</pre>
If you don't have wxWidgets, or your current version is too old,
the following invocation should build
a wxWidgets distribution for you. (Be warned that this may take
some time.)
<pre>
../configure --enable-buildwx
</pre>
</p>
<p>If none of these variations for invoking the configure
script worked for you, skip ahead to
<a href="#advanced">Advanced Configuration Options</a>.
</p>
<p>
You are now ready to make lamarc. (It is recommended that you use
the GNU make utility. On some systems it may be called gmake.)
<pre>
make
</pre>
This will produce executables lamarc and lam_conv if you're running
on a Unix or Linux system. If you're running under Mac OS X, you
should get clickable executables.
</p>
<p>
To install the executables and html documentation (you may
need sysadmin privilidges to do this, and it is not required
to use the program) type either
<pre>
make install
</pre>
or
<pre>
make install-strip
</pre>
</p>
<p>
Either of these should install the executables in /usr/local/bin and
the documentation under /usr/local/html/lamarc. Executables installed
either way have the same compute time performance. The install-strip
versions are smaller but provide no debugging information, should
you run into problems.
</p>
<h3><a name="advanced">Advanced Configuration Options</a></h3>
<p>
For most users,
the configure script provided with the distribution should work
as <a href="#basic">described above</a>.
However, there are several situations in which you may wish
to change the behavior of the configure script. These include
when:
<ul>
<li>the configure declares it <a href="#ostype">cannot proceed without
LAMARC_CODE_OSTYPE</a>,</li>
<li>the configure declares it <a href="#gui">cannot proceed without
GUI_TOOLKIT</a>, or</li>
<li>you wish to <a href="#special">use a non-default compiler, or override
variables such as CXXFLAGS and LDFLAGS</a>.</li>
</ul>
</p>
<p>
If these techniques are not adequate to solve your problems, you
may wish to:
<ul>
<li>compile a <a href="#batchOnly">batch only version of the lamarc converter</a>,</li>
<li>default to an <a href="#ancientConverter">ancient version of the lamarc converter</a>, or</li>
<li><a href="#selfEdit">edit the configure script and Makefile yourself</a>.</li>
</ul>
</p>
<h4><a name="ostype">Setting LAMARC_CODE_OSTYPE</a></h4>
<p>
While most of the lamarc code is not dependent on the user's operating system,
there are a few minor differences in I/O handling and standard library headers.
These differences require the setting of the variable <b>LAMARC_CODE_OSTYPE</b>
so that appropriate code can be exercised for each operating system.
</p>
<p>
Under normal circumstances, the configure script should be able to
guess the correct value for LAMARC_CODE_OSTYPE. If it cannot, you
may set the value by including an assignment to it when you
invoke the configure script.
For example
<pre>
../configure LAMARC_CODE_OSTYPE=LAMARC_COMPILE_MACOSX
make
</pre>
The assignment to LAMARC_CODE_OSTYPE should come after any other arguments
to the configure script, for example:
<pre>
../configure --enable-debug LAMARC_CODE_OSTYPE=LAMARC_COMPILE_LINUX
make
</pre>
Legal values for LAMARC_CODE_OSTYPE are:
<ul>
<li>LAMARC_COMPILE_LINUX</li>
<li>LAMARC_COMPILE_MACOSX</li>
<li>LAMARC_COMPILE_WINDOWS</li>
</ul>
</p>
<h4><a name="gui">Setting GUI_TOOLKIT</a></h4>
<p>
Under normal circumstances, the configure script should be able to
guess the correct value for GUI_TOOLKIT. If it cannot, or you wish
to use a different toolkit, you should set the value for GUI_TOOLKIT.
As with <a href="#ostype">setting LAMARC_CODE_OSTYPE</a>, the assignment
is made at the end of the invocation of configure.
For example:
<pre>
../configure GUI_TOOLKIT=gtk2
make
</pre>
The assignment to GUI_TOOLKIT should come after any other arguments
to the configure script, and is most likely to be needed when
you are also setting LAMARC_CODE_OSTYPE. For example:
<pre>
../configure --enable-debug LAMARC_CODE_OSTYPE=LAMARC_COMPILE_MACOSX GUI_TOOLKIT=mac
make
</pre>
</p>
<p>
Tested values for GUI_TOOLKIT are:
<ul>
<li>gtk2 -- tested on linux systems</li>
<li>mac -- tested on OS X systems</li>
<li>msw -- tested on a Windows system</li>
</ul>
It is possible that other toolkit values supported by wxWidgets work,
including x11, Motif, and OS/2. If you wish to build for one of
these toolkits, we suggest you start with wxWidget's
<a href="http://wiki.wxwidgets.org/Supported_Platforms">
information on supported platforms
(http://www.wxwidgets.org/docs/platform.htm)</a>.
</p>
<h4><a name="special">Specifying non-standard compiler and/or build flags</a></h4>
<p>
The configure script is designed to allow you to
specify a compiler or to pass options to the compiler
or preprocessor. For example, if you are testing out
a new C++ compiler you might invoke it like this:
<pre>
../configure CXX=/path/to/my/compiler CXXFLAGS="I /path/to/my/include/files"
make
</pre>
</p>
<p>
For more information on options and environment
variables that effect the configure script type
<pre>
../configure --help
</pre>
</p>
<h4><a name="batchOnly">Compiling a batch only version of the converter</a></h4>
<p>
If you cannot successfully build for any GUI_TOOLKIT,
or if you wish to run the lamarc file converter as
part of a batch process, you can build a batch only
version of the converter with the following commands
(add CXX and CXXFLAGS, etc at the end if you need them):
<pre>
../configure --disable-gui --enable-buildwx
make
</pre>
</p>
<h4><a name="ancientConverter">Compiling a wxWidgets-less converter</a></h4>
<p>
We recommend the following step only as a <b>last resort</b>.
If you are going to attempt it, we recommend that you
email us at
<A HREF="mailto:lamarc@u.washington.edu">lamarc@u.washington.edu</A>.
to make sure there isn't a better solution.
</p>
<p>
If you unable to build wxWidgets at all (and therefore not
have access to the converter) you may instead configure and make as
follows:
<pre>
../configure --disable-converter
make old_lam_conv
make
</pre>
You must explicitly give the argument to "make old_lam_conv" --
it will not otherwise be built.
Please be warned that <b>old_lam_conv</b> is <b>no longer maintained</b>
and will not have all features of the new converter.
</p>
<h4><a name="selfEdit">Editing the configure script and Makefile yourself</a></h4>
<p>
If none of the above techniques allows you to build the lamarc
programs, you have one further option -- edit the configure
script and Makefiles yourself.
</p>
<p>
The easiest way to do this is to edit them the same way we do,
using autoconf and automake.
These tools are available at the following locations:
<ul>
<li><a href="http://www.gnu.org/software/autoconf/">http://www.gnu.org/software/autoconf/</a>
<li><a href="http://www.gnu.org/software/automake/">http://www.gnu.org/software/automake/</a>
</ul>
</p>
<p>
You may also wish to read additonal documentation
available at
</p>
<ul>
<li><a href="http://sources.redhat.com/autobook/">http://sources.redhat.com/autobook/</a>
<li><a href="http://seul.org/docs/autotut/">http://seul.org/docs/autotut/</a>
</ul>
<p>
Make any edits you need to Makefile.am or configure.ac
and then issue the following series of commands
</p>
<pre>
aclocal
autoconf
autoheader
automake --add-missing
</pre>
<p>
You may then proceed to configure and make as in the section
<a href="#basic">Basic Instructions...</a> above.
</p>
<P>(<A HREF="mapping.html">Previous</A> | <A
HREF="index.html">Contents</A> | <A HREF="converter.html">Next</A>)</P>
</BODY>
</HTML>
|
